About the book
Geostatistical modeling too often falls into the trap of "button pushing" on commercial software without an understanding of the basic underlying principles. After more than 30 years of teaching earth modeling classes, it is clear that the base knowledge of geostatistical principles has grown amongst earth modelers, yet, there remains a great deal more to learn. Many modelers today have become lost in software products and rely too heavily on embedded defaults, or suggestions from colleagues.
This book is intended to be a companion to modelers interested in knowing the practical meaning of what is behind the buttons they are pushing. It is not a textbook on the mathematics of geostatistics or the evolution of its theory. It is a guide to help make practical decisions and simply explain the “why” and “how” of what works and what does not. Further, it will attempt to answer questions where difficult choices and resulting implications are not clear; e.g. What variogram model should I use? What simulation algorithm is best? How many realizations should I run? After a review of basic principles and common pitfalls, case study examples will be drawn from both conventional and unconventional reservoirs. The case studies will be followed by a constructive review from a panel of experts geostatisticians articulating both strong and weak points of the models, and offer suggestions. Finally, the impact of high-performance computing, machine learning, data analytics (big and small), Python, and R will be discussed with a view towards successful earth modeling for the next decade.